Brown 1993.

Methods	Multicentre RCT (4 centres)
Participants	OCTR (85 hands in 75 participants) or 2‐portal ECTR (84 hands in 76 participants). (Some participants had OCTR in one hand and ECTR in the other) Average age 55 years (range 25 to 87 years); 99 women, 46 men. The dominant hand was involved in 104 participants. The average duration of the symptoms before the operation was 25 months (range 2 months to 10 years). Diagnosis was clinical and confirmed electrophysiologically All participants had either had failure of a trial of non‐operative management with a wrist splint, steroid injections into the carpal canal or both, or they had refused such a program. The duration of pre‐operative treatment ranged from no treatment to 10 years
Interventions	2‐portal ECTR (extrabursal) vs OCTR
Outcomes	Follow‐up at 3 and 6 weeks and 3 months Relief of numbness and paraesthesias, satisfaction with the procedure, interstitial pressures in the carpal canal; 2‐point discrimination, Semmes‐Weinstein monofilament, motor strength, electrophysiological testing and functional outcomes (grip strength, key pinch strength, tenderness of the scar, pillar pain, recovery of the ability to perform activities of daily living, and return to work)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	After the decision to proceed with CTR had been made, a random draw was done to determine which procedure a participant would have. Slips of paper (2 cm2), labelled either group I or group II, were placed in an opaque hat; the hat was lifted above eye level; and a single slip of paper was chosen by a member of the operative team. Group I participants were assigned to standard OCTR and group Il participants to modified 2‐portal ECTR
Allocation concealment (selection bias)	High risk	After the decision to proceed with CTR had been made, a random draw was done to determine which procedure a participant would have. Slips of paper (2 cm2), labelled either group I or group II, were placed in an opaque hat; the hat was lifted above eye level; and a single slip of paper was chosen by a member of the operative team. Group I participants were assigned to standard OCTR and group Il participants to modified 2‐portal ECTR
Blinding of participants and personnel (performance bias) All outcomes	High risk	No information given. Participants and personnel could not be blinded
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The participants were evaluated by an independent observer who was blinded as to which operation had been done. At each centre, one certified hand therapist always performed this task. All of the questions and evaluation techniques were standardised to minimise interobserver variability. A stockinette with occlusive 10 cm × 10 cm cotton gauze was placed over the participant’s wrist and palm
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	"One hundred and forty‐nine hands (88 per cent) had follow‐up at twenty‐one days; 147 hands (87 per cent), at forty‐two days; and 160 hands (95 per cent), at eighty‐four days."
Selective reporting (reporting bias)	High risk	Not all the prespecified outcomes at all time points were reported
Other bias	Low risk	"No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study." No baseline imbalance